The NVIDIA Jetson T5000™ module gives you the highest level of performance for physical Al and robotics. It delivers up to 2070 FP4 TFLOPS of Al compute and 128 GB of memory with power configurable between 40 W and 130 W. Compared to AGX Orin™, it delivers over 7.5x higher Al compute than NVIDIA AGX Orin", with 3.5x better energy efficiency.
This system-on-module (SOM) features an NVIDIA Blackwell architecture GPU with a transformer engine and Multi-Instance GPU (MIG) support to effortlessly run the latest generative Al models. NVIDIA Jetson Thor™ accelerates low-latency, real-time multi-sensor applications with a 14-core Arm Neoverse-V3AE CPU, 4x 25 GbE networking, and extensive I/O options for sensor fusion. It also includes a suite of accelerators, including a third-generation Programmable Vision Accelerator (PVA), dual encoders and decoders, an optical flow accelerator, and more.
Jetson Thor belongs to a new class of robotic computers, architected from the ground up to power next-generation physical Al applications. It supports a wide range of generative Al models, from Vision Language Action (VLA) models like NVIDIA Isaac™ GROOT for humanoids to all popular LLMs and VLMs. To deliver a seamless cloud-to-edge experience, Jetson Thor runs the NVIDIA Al software stack for physical Al applications, including NVIDIA Isaac for robotics, NVIDIA Metropolis for visual agentic Al, and NVIDIA Holoscan for sensor processing. You can also build Al agents at the edge using NVIDIA agentic Al workflows like Video Search and Summarization (VSS).
Our ecosystem of partners offers all the carrier boards, design services, cameras, and other sensors you need, as well as additional Al and system software. This lets you accelerate solution development in industries ranging from robotics and smart spaces to retail, industrial, medical, and more.
The Jetson Thor module gives you production-ready performance, massive Al compute, and sensor capabilities for physical Al applications in a compact form factor. This makes it the ideal platform for developers looking to unlock new possibilities for humanoid robotics and other physical Al applications.
Key Features
NVIDIA Jetson T5000 Module
2070 FP4 TFLOPS AI Compute powered by Blackwell GPU
4-core ARM® Neoverse® CPU
128 GB LPDDR5X Memory
Explore full Jetson Thor documentation and resources — visit NVIDIA Developer.
Technical Specifications
Category
Specification
Product
NVIDIA Jetson T5000
AI Performance
2070 TFLOPS (FP4—sparse)
GPU
2560-core NVIDIA Blackwell architecture
GPU with 96 fifth-gen Tensor Cores
Multi-Instance GPU with 10 TPCs
GPU Max Frequency
1.57 GHz
CPU
14-core Arm® Neoverse®-V3AE 64-bit CPU;
64 KB I-Cache, 64 KB D-Cache;
1 MB L2 Cache per core;
16 MB shared system L3 cache
CPU Max Frequency
2.6 GHz
Vision Accelerator
1× PVA v3
Memory
128 GB 256-bit LPDDR5X,
273 GB/s
Storage
Supports NVMe through PCIe;
Supports SSD through USB 3.2
Video Encode
6× 4Kp60 (H.265);
12× 4Kp30 (H.265);
24× 1080p60 (H.265);
50× 1080p30 (H.265);
48× 1080p30 (H.264);
6× 4Kp60 (H.264)
Video Decode
4× 8Kp30 (H.265);
10× 4Kp60 (H.265);
22× 4Kp30 (H.265);
46× 1080p60 (H.265);
92× 1080p30 (H.265);
82× 1080p30 (H.264);
4× 4Kp60 (H.264)
Camera
Up to 20 cameras via HSB;
Up to 6 cameras through 16× lanes MIPI CSI-2;
Up to 32 cameras using virtual channels;
C-PHY 2.1 (10.25 Gbps);
D-PHY 2.1 (40 Gbps)
PCIe*
Up to 8 lanes—Gen5;
Root port only—C1 (x1) and C3 (x2);
Root Point or Endpoint—C2 (x1), C4 (x8), and C5 (x4)
